Oil sources administered to tambaqui (Colossoma macropomum): growth, body composition and effect of masking organoleptic properties and fasting on diet preference

Tropical fish feeding behaviour and food intake are regulated by a number of environmental factors (e.g.: stress in intensive aquaculture; type or seasonality of food), and also by complex homeostatic mechanisms that involve nutritional requirements. When access to food is not restricted, e.g. controlled laboratory conditions, fish growth and behaviour remain relatively unaffected. Moreover, fish need dietary fatty acids and, while some are remarkably beneficial, such as long-chain omega-3 oils, other fats can be associated with poor health, mainly under adverse conditions. This study aimed to evaluate the performance and body composition of tambaqui juveniles fed different oil sources (fish, linseed, and corn) (Experiment 1) and the influence of post-absorptive signals or orosensorial properties of the dietary oil source on diet preference (Experiment 2). For Experiment 1, juvenile tambaqui (42.79 ± 0.92 g) were placed into eighteen 100 L tanks (10 fish per tank). Three diets (treatments) were formulated with three oil sources (corn, linseed and fish oil). Three groups of fish were fed each separate diets twice a day for 7 weeks. Experiment 2 assessed tambaqui's ability to select oil sources without coming into contact with the sensorial properties of diet. The same diets of Experiment 1 were used, but diets were encapsulated to isolate sensorial properties. Forty-eight fish (41.26 ± 1.00 g, mean ± SD) were distributed into six 250 L tanks (eight fish per tank). In experiment 1, we found no effects of diet on growth parameters, such as final body weight, weight gain, feed intake, feed efficiency ratio, specific growth ratio and protein efficiency (p ≥ 0.05). However, body crude protein was higher in the fish fed the corn oil diet (p ≤ 0.05), but this value did not affect the protein efficiency ratio (p ≥ 0.05). No differences among treatments were observed for moisture, lipid and ash (p ≥ 0.05). The fatty acid composition of muscle reflected the dietary oil source. However, the tambaqui fed vegetable oil displayed high levels of polyunsaturated fatty acids, while the linseed oil diet resulted in a greater incorporation of n-3 highly unsaturated fatty acids. In Experiment 2, at first fish did not show the ability to select an oil source without orosensorial properties, but a nutritional challenge of food deprivation (10 days) enabled fish to select among the diets that contained fish oil sources (p = 0.002) by using post-ingestive signals. Our findings contribute to the understanding of the mechanisms that relate to lipid metabolism and feeding behaviour in freshwater Amazon species, which goes beyond growth parameters.